International Journal of Accounting and

Financial Management Research (IJAFMR)

ISSN(P): 2249-6882; ISSN(E): 2249-7994
Vol. 5, Issue 2, Apr 2015, 1-16
TJPRC Pvt. Ltd.

WHEN IS ACCOUNTING QUALITY IMPROVED? ASSESSING THE

EARNINGS MANAGEMENT EFFECT OF IFRS FOR SOUTH AFRICA
B ENJ AMIN YEBOAH1 & MICHA EL YEB OAH2
1

PhD Student, ISCTE Business School, Lisbon, Portugal, Ku masi Polytechnic,

Depart ment of Accounting and Accounting Information Systems, Ku masi, Ghana

2

Lecturer, Ku masi Po lytechnic, Ku masi Polytechnic, Depart ment of Accounting and

Accounting Information Systems , Ku masi, Ghana

ABSTRACT
The study tests the extent of the IFRS adoption by South Africa listed firms as a t 1st January, 2005 in relation to
earnings management on accounting quality for spanning period between 1998 and 2012. The dimensions of earnings
management that were analyzed are a change in net income, variability of change in net income over change in cash flow
fro m operations, correlation between accruals and cash flow, change in net income over small positive target (SPOS), and
absolute discretionary accruals. The study employed the ordinary least square (OLS) estimators based on prior studies to
measure the metrics of earnings management. We find that the adoption of IFRS ha s resulted in better accounting quality
than previously under South African Generally Accepted Accounting Principles (GAAP). Specifically, the results evidence
that the pervasiveness of earnings management by way of earnings smoothing and discretionary accruals has reduced
within post adoption period. This conclusion is contrary to SPOS metric.

KEYWORDS: IFRS, Accounting Quality, Earn ings Management, Capital Market, South Africa
1 INTRODUCTION
The change in accounting regimes to the adoption of International Financial Reporting Standards (IFRS) is
marked as one of the most significant in accounting history. Higher accounting quality, and the likes of improved corporate
transparency and improved liquidity position, to benefit firms and investors (Barth et al. 2008; Armstrong et al. 2010;
Zghal et al. 2011) are attributed to the IFRS adoption. South Africa mandated IFRS adoption in 2005 as like many
European Union (EU) countries.
There is increasingly of corporate accounting discredit among corporate leaders and professional accountants in
South Africa as accounting fraud is the order of the day, due to relatively weak institutional environ ment. In view of that
the earnings management are under severe attack by corporate media and financial analysts. Many accounting scholar are
seemingly against the practice of earnings management, even though it is accepted that not all earnings management
techniques are deceptive. This study therefore examines the effect of earnings management incentives on accounting
quality of South Africa listed firms.
Practice of earnings management and its adverse consequences on the financial reporting process are predominant
and a source of worry to the users of accounting information in South Africa. Many scholars have already conducted
extensive research about the extent of earnings management on the accounting quality of publicly European Union traded

www.tjprc.org

editor@tjprc.org

Benjamin Yeboah & Michael Ye boah

firms (Ba rth et al. 2008). Yet, to our knowledge, there is little empirical evidence on the impacts of the earnings
management on accounting quality after the adoption of IFRS, especially fro m developing countries perspective.
Therefore, by using a firm-level cross-sectional regressions and discretionary accruals models, this study aims to exploit
the extent of mandatory IFRS adoption and the earnings management on the quality of accounting information of the JSE
firms.
South Africa is first among the countries in Africa outside the EU to have mandated the adoption of IFRS.
Therefore, this paper contributes to the existing literatures that have largely focused on only EU adoption. The study
contributes to the growing body of literature on earnings management of capital markets of Africa. Furthermo re, South
Africa has a comparat ively longer adoption experience relative to other African countries that mandated the post 2005
IFRS adoption, and therefore allows a sufficient informat ion window to assess such impact, since it has enough time -frame
for meaningfu l conclusions.
We use a wide range of accounting-based metrics and discretionary accruals for comparing the quality of
accounting numbers under South African GAAP and IFRS . First, we co mpare the pervasiveness of earnings management
under South African GAAP and IFRS by examin ing the extent in which earnings are smoothed and managed towards a
positive target. Second, we assess whether the IFRS adoption has affected the discretionary accruals on listed firm of JSE
(Jeanjean & Stolo wy, 2008). The evidence reveals that IFRS adoption in South Africa leads to a significant improvement
in accounting quality in all metrics emp loyed, except the small positive net inco me towards target (SPOS).
The remainder of this paper is organized as follows. The next section reviews the re levant literature on the
adoption of IFRS and earnings management. The third section exp lains the research design and sample data employed in
the study. The fourth section presents the descriptive and empirical results, and we provide our conclusion in the final
section.

2. LITERATURE REVIEW
a. Accounting Practices Board and IFRS
The South African Institute of Chartered Accountants (SAICA) as well as Accounting Practices Board (APB)
promu lgates South African accounting standards. Early 1995, SAICA has b een gradually adopting IFRS but with minor
modifications (Prather-Kinsey, 2006). Furtherance to the total harmonizat ion with IFRS, South Africa principles were
brought to bear with IFRS in effect ive 2000. This basis requires the firms that are listed in the Johannesburg Securities
Exchange (JSE) to only present their financial statements in accordance with IFRS as fro m 1 st January, 2005.
b. Earnings Management Defi nition
Schipper (1989) defines earnings management as a purposeful intervention in the external financial reporting,
with the intent of obtaining some private gain. In consistent with Schipper (1989), Mu lford and Co miskey (2002) defined
earnings management as the active manipulation of earnings toward a predetermined target.
c. Earni ngs Management Incenti ves
Broadly, the study highlights three main incentives for earnings management as identified by prior studies (Healy
& Wahlen, 1999):

Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0

When is Accounting Quality Improved? Assessing the Earnings Management Effect of IFRS for South Africa

Capi tal Market Incenti ves

To avoid a significant decline of co mpanys share price, managers man ipulate earnings and future sales in order
to achieve earnings above benchmarks. The incentive to do that is also heightened when top managers decide to realign
their remunerat ion packages like stock options and bonus with firm performance (Matsunaga & Park, 2001). In most cases,
they engage on this exercise to receive h igher prices for their stock.
Contractual Moti ves
The incentives for earnings manipulation of this motive involves a situation where top managers manipula te
earnings to gain more earnings for meeting their legal contractual obligations with debt lenders, suppliers, and creditors
when fall due. This exercise creates room for more earn ings and avoids losses to meet such obligations so as to avoid
penalty. This motive is heightened within the synopsis of agency theory.
Regulatory Moti ve
This motive exp lains the political cost hypothesis as proposed by Watts and Zimmerman (1978). It exp lains that
big firms (size) that report high earnings can draw the attention of regulators and the government in terms of implement ing
political actions to avoid discrediting the ruling government. In an attempt to oppose such actions would lead to political
actions such as legal costs against the firm. Moreso, wage increase demand by labour unions may occur, base on large
reported earnings. Consequently, large firms have incentives to decrease the amount of reported earnings.
d. Accounting Quality and Earnings Management
There are a lot of corporate scandals (e,g Enron, Parmalat) in the global world that explain and attribute its cause
to the phenomenon of earnings management (Goncharov, 2005). Accounting quality is a fundamental basis for researching
and is the motives and consequences of financial reporting manipu lation s within financial statements . This basis has
necessitated indepth investigations into the existence and practice of earn ings management by accounting researchers . In
view of that, accounting quality becomes questionable when managers have an incentive to manage reported earning s
when the need arises (Dechow & Skinner, 2000).
Payne and Robb (2000) study the effect of analysts' forecasts on earnings management. They conclude that
managers have incentives to use their discretion over accounting accruals to eliminate negative earnings surprises.
They find that managers manage earnings upwards when pre-managed earnings are below analysts' forecasts. However,
when pre-managed earnings exceed analysts' forecasts, managers either keep discretionary accruals for future periods by
emp loying an inco me-decreasing strategy or preserve a positive earnings surprise with the aim of ach ieving a favourable
share price reaction. Holland and Ramsay (2003) examine whether listed Australian companies manage earnings to report
positive profits. They find evidence of discontinuities in the distribution of reported earnings and changes in earnings.
Sarkar et al (2006) ascribe low levels of earnings management to a situation of high quality governance mechanisms, such
as independent board of directors and effective existence of audit committee. They investigated the impact of board
characteristics on opportunistic earnings management in the context of a large emerging Ind ian economy, using a sample
of 500 large Indian firms. The results indicated that it is not board independence per se, but rather board quality that is
important for earnings management. The results showed that diligent boards and audit committee are associated with lower
earnings manipulat ion.

www.tjprc.org

editor@tjprc.org

Benjamin Yeboah & Michael Ye boah

2.1 Hypothesis
South African financial reporting environ ment achieves a co mpetitive advantage arising fro m early IFRS adoption
over other African countries. In addit ion, there is existence of high-quality national accounting regime that is well-regarded
before transition to IFRS. This imp lies that the country had already enjo yed high-quality reporting practices prior to the
introduction of IFRS. Despite this, prior few researches about the IFRS adoption and earnings management on accounting
quality conclude mixed findings. Some research work attributes to the fact that South A fricas former accounting
principles-based standards are similar to the use of IFRS, and therefore its effect is negligible. Th is makes the current IFRS
adoption effect to be still uncertain.
The regulators of the standards predict that investors should b enefit from the adoption of IFRS, since it ensures a
high

degree

of

transparency

and

comparability

of

financial

statements

for

quality

financial

report ing

(EC Regulation No. 1606/2002). Following this assertion, it is expected that the adoption of IFRS would limit management
ability to have opportunity to exercise their own discretion over certain key accounting decisions to their advantage.
This tends to provide more consistent approach to accounting measurement , which leads to improve accounting quality
(Daske et al, 2008). Despite that, there is another school of thought which argues that earnings management has not
improved accounting quality, especially in the short timeframe after IFRS adoption (Jeanjean & Stolowy, 2008). Ewert and
Wagenhofer (2005) study conclude that tightening the accounting standards can reduce the level of earnings management,
thereby imp roving the quality of accounting.
Taken together the general underlying concept of IFRS adoption and mixed findings of accounting quality arising
fro m the adoption in developing countries, we propose this hypothesis;
HI: The adoption of IFRS will lead to higher accounting quality in the post-adoption period arising fro m less
earnings management in South Africa.

3. RESEARCH METHODS AND PROCEDURES

a. Type of Business Research
The studys research type is quantitative method. We used this method to collect secondary data from the
DataStream database. Among the financial statement indicators collected includes relevant income statement items,
financial position items and other indicators for the study. In view of inconsistencies of the South African data, a matched
sample construction was not employed. The accounting quality metrics applied in the study of earnings management were
earnings smoothing and discretionary accruals.
b. Data Collection and Sample Size
Our samp le co mprises 2535 firm-year observations for 181 firms that adopted IFRS between 1998 and 2012 for
which DataStream data are available for the period understudy. The sample excludes debt instruments firms and mutual
companies in South Africa. We gather both stock data and accounting data from DataStream. The population in this study
refers to all co mpanies listed in the JSE between 1998 and 2012. The need for this study is appropriate since p rior works
never considered such an extensive period baseline study. The sample excludes fund firms, debt firms, dead firm with in the
study period, and firms with missing data. The sample firms cut across all forms of industry such as extraction, finance,
manufacturing, merchandise, and service firms as adapted from Barth et al. (2009) classification.

Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0

When is Accounting Quality Improved? Assessing the Earnings Management Effect of IFRS for South Africa

All the accounting data were expressed in U.S Dollars. The study focused on listed companies in South Africa,
where comp liance of IFRS is mandatory and therefore data was of high level of integrity. Consistent with the European
Union (EU), the South African transition to IFRS was made effective fro m January, 2005, and therefore would have
expected the first IFRS accounts in the fiscal year-ended December 31, 2005. In line with Paananen and Lin (2007), some
institutional factors held constant including stock listing requirements, accounting disclosure requirements, market
microstructure and regulatory environments. This tends to strengthen the reliability of the findings thereon. In view of
inconsistencies of the data, a matched sample construction was not employed.
c. Measurement of Accounting Quality
We mostly used two prominent measures that are mostly employed by prior researchers in measuring for
accounting quality during the IFRS adoption to include: earnings management and discretionary accrual methods
(Barth et al, 2008; Kothari et al, 2005; Jones et al, 2008). The effects of financial reporting system are the basis for metric
of accounting quality. In addition to this, there are suitable economic and environmental conditions that serve as incentives
to be associated with the adoption of IFRS. The study intended to overcome these challenges and mit igate the confounding
factors to ensure complete and reliable outcome. We assess the impact of IFRS adoptions on accounting quality between
the pre-adoption (1998-2004) and post-adoption (2006-2012) periods. The study is based on listed companies in the
Johannesburg Stock Exchange (JSE) in South Africa, where adoption comp liance is mandatory.
Contrary to Lang et al (2003), the study never employed a paired sample, but considered two main divisions
between the pre-adoption period and post-adoption period of the South African GAAP and IFRS respectively.
d. Earnings Management Measurement Methods
This measure employs four earnings management metrics to examine accounting quality, namely: earnings
smoothing, managing earnings towards targets, and discretionary accruals. Specifically earnings smoothing recognises
metrics such as: the variability of the change in net income, the variability of change in net income over the variability of
change in cash flow fro m operations , managing earnings towards targets and the other one is discretionary accruals .
1. Earnings Management
It is noted that earnings management measurement a challenging task for researchers , due to the fact that earnings
management is invisible to detect especially for s mall data s ize, and therefore it is not easy for effective management.
The study follows Barth et al (2008) three prominent measures of earnings management namely : the variability changes in
earnings, variability of changes in net income earnings over variabil ity in change in cash flow, the spearman correlat ion
between accruals and cash flow, managing towards small positive earnings, and the last construct is the absolute
discretionary accruals. These methods are discussed in turn as:
a. Variability in Change in Net Earnings
Barth et al (2008) explains the above measure as a small variance of the change in net income is interpreted as
evidence of earnings smoothing. This construct is affected by other unrelated factors to earnings smoothing. The need to
mitigate against these factors recognize the measure of earnings variability as the variance of the residuals from the
regression of change in net inco me. In the equation (1), the variab ility of net inco me in natural logarithm is between pre
and post periods of the IFRS adoption of the regression against the control variab les.

www.tjprc.org

editor@tjprc.org

Benjamin Yeboah & Michael Ye boah

Equation (1): Regression of LCg NI on the control variab les

LCgNI it 0 1 SIZEit 2 CFOit 3 LGROWTH it 4 LEVit 5TURNit 6 ISSUEit 7 INDi 8TIMEi it

Where,
SIZE =the natural log of total assets
CFO = annual net cash flow fro m operating activit ies divided by total assets
LGROWTH = the natural log of percentage change in sales
LEV = total liabilit ies divided by the year end equity book value
TURN = sales divided by year end total assets
ISSUE = percentage change in total common stock
IND = variables for industry fixed effects, being classified
TIM E = du mmy variables for t ime (year) fixed effects
b. Vari ability of Change in Net Income Over Change in Cash Flow
On the general basis, firms that have more volatile cash flow typically , have a volatile net inco me, especially
when firms use accruals to manage earnings. This condition tends to lower variation in inco me than that of operating cash
flow. The model for equation (2) specifies this as:
Equation (2 ): Regression of LCg CFO on the control variables
LCgCFOit 0 1 SIZEit 2 LGROWTH it 3 LEVit 4TURNit 5 CFOit 6 ISSUEit 7 INDi 8TIMEi it

c. Spearman Correlati on of Accruals and Cash Flow

Spearman correlation between accruals (ACC1) and cash flows (CFO) is the third earnings smoothing measure.
When firms attain poor cash flows, firms use accruals to engage in earnings management to smooth cash flow variab ility.
Consistent with the previous measures (Barth et al. 2008), other factors could similarly influence cash flow and accruals.
We use residuals fro m regressions of cash flows and accruals to determine the spearman correlation. Equation (3a)
addresses the natural log of cashflow metric, wh ilst equation (3b) takes care of accruals (ACC1) metric.
Spearman correlation between cash flows and accruals
= CORR (CFO Rho, A CCI Rho)
Where: CFO = residuals fro m the regression of CFO on the control variables
ACCI =residuals fro m the regression of ACCI on the control variab les
Equation (3 a): Regression of CFO on the control variab les

LCFOit 0 1 SIZEit 2 LGROWTH it 3 LEVit 4TURNit 5 ISSUEit 6 INDi 7 TIMEi it

Equation (3 b): Regression of ACC1 on the control variab les

Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0

When is Accounting Quality Improved? Assessing the Earnings Management Effect of IFRS for South Africa

ACCI it 0 1 SIZEit 2 LGROWTH it 3 LEVit 4TURNit 5 ISSUEit 6 INDi 7TIMEi it

d. Management towards small positi ve earnings (SPOS)
This construct follows Lang et al (2006) and Barth et al (2008) by using a dummy variable that set to 1 for
observations for which annual net income, scaled by total assets is between 0 and 0.01, and sets to 0 otherwise. We run the
logistic regression by denoting STDS as binary variable of SPOS, the dependent variable. This construct examine whether
the probability of firms reporting small positive earnings (SPOS ) has changed after firms transited to IFRS, together with
control variables used by previous study (Barth et al, 2008).
Equation (4 ): Log istic regression of SPOS on STDS and the control variables
SPOSit 0 1 STDit 2 SIZEit 3CFOit 4 GROWTH it 5 ACCI it 6 LEVit 7TURNit 8 ISSUEit 9 INDi 10TIMEi it

Where:
STDS =du mmy variab le that equals 1 if observation are in the post adoption, and 0 otherwise
SPOS= dummy variable that equals 1 if net income scaled by total assets is between 0 and 0.01, and 0 otherwise
2. Discretionary Accruals Methods
Discretionary and non-discretionary accruals are used to measure managed and unmanaged earnings respectively.
As pointed out by Healy (1985), management exercises discretion over discretionary accruals only when managing
earnings. Earlier studies examine specific choice of accounting methods and accounting estimates managers made under
discretionary accruals for measuring earnings (McNichols and Wilson, 1988). Total accruals are defined as the difference
between cash from operations and net income (DeAngelo 1986; Healy, 1985) . Several emp irical studies show that
accrual-based accounting earnings are more informative with respect to stock returns than cash flows (e.g. Dechow, 1994).
Discretionary accruals allow managers to exercise their d iscretion over accounting choices and estimates, and therefore a
platform to promote earnings management (e.g Dechow et al, 1995; Batov et al, 2001). A mong the most commonly
methods used by previous studies to measure earnings management are abridged in Table 1.
Table 1: Discretionary Accrual Methods for Measuring Earnings Management

www.tjprc.org

editor@tjprc.org

Benjamin Yeboah & Michael Ye boah

The Jones (1991) model regresses total accruals on gross property, plant and equipment and changes in revenues
which provide coefficients that are then used to estimate un managed accruals. One major limitation of this model lies in
the assumption that managers do not exercise discretion over revenues and this can lead to misspecification of the
discretionary accruals when managers do exercise discretion over revenues. Several emp irical studies pinpoint that
Dechow et a1.s (1995) modified Jones model is best in detecting earnings management, specially using it on a cross
sectional discretionary accruals basis (e. g; Jiang et al, 2008; Chang & Sun, 2009).
Kothari et al. (2005) argue that the discretionary accruals, as estimated by both Jones and modified Jones models
may result in severe measurement error in discretionary accruals when these models do not control for the prior
performance of the co mpany. They propose a model that includes control for the firms performance using the lag of return
on assets (ROA) to mitigate the problematic heteroskedasticity and mis specification issues of the Jones and modified Jones
models in estimating accruals. The non-discretionary accruals of the cross -sectional modified Jones model with current
year ROA is what this paper used as estimated in equation (5):

NDAi ,t 1 (1/ Assetsi ,t 1 ) 2 (REVi ,t RECi ,t ) 3 PPEi ,t 4 ROAi ,t

Equation (5)

The definitions of the variables are the same as for Table 1. The estimates of the industry-specific parameters are
1, 2, 3, and 4 .
3.1 Measuring Total Accruals
There is the need to compute total accruals first in order to estimate discretionary accruals. The literature offers
two methods for computing total accruals. The first is the traditional balance sheet approach that is used in the majority of
prior studies (e.g. Dechow et al.; 1995; Peasnell et al.; 2000b; Kothari, 2001). The second method is the cash flow
approach used by recent studies (e.g. Huang et al, 2007). One of the reasons for the popularity of the balance sheet
approach may be the availability of balance sheet statement data compared to cash flow statement data. This study
emp loyed the cash flow method due to availability of relevant data for computation. The formu la for cash flo w approach
measures accruals as: ACCI =NI CFO, where ACCI is total current accruals, NI is earn ings before interest and tax, and
CFO is cash flow fro m operations (Jaggi et al, 2009), whileas the balance sheet approach follows Dechow et al. (1995) and
Kothari (2001) prior methods as shown in equation (6):

TACt CAt Casht CLt DLt Dept

Equation (6)

Where:
CA t = Change in current assets in year t
Cash t = Change in cash and cash equivalents in year t
CLt = Change in current liabilities in year t
DCLt = Change in debt included in current liabilities in year t
DEPt = Depreciation and amortizat ion expense in year t
Given the above formula fo r both the total current accrual (ACC1) and cash flow (CFO) perspective and
non-discretionary accruals (NDA), the discretionary accruals (DA) (See equation 5) can be derived as:

Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0

When is Accounting Quality Improved? Assessing the Earnings Management Effect of IFRS for South Africa

DA = A CC1 NDA

Equation (7)

This is where DA is used as a proxy for derivation of earnings management computation. Alternatively, the
residuals of NDA (see equation 5) could be employed as a proxy for earnings management serving as a dependent variable
(Morais & Curto, 2009; Sam et al, 2010). For the purpose of this study, we employed the residuals of NDA approach for
estimating the DA (see Tables 5 and 6a and 6b). Collins and Hribar (2002) find that the balance sheet approach has a high
frequency and magnitude of significant errors, it is less efficient than cash flow approach.

4. RESULTS
1. Descripti ve Statistics
The descriptive statistics in Table 2 relates to variables the study used in our analyses. It shows that the
post-adoption firms have significantly fewer incidences of small positive earnings and significantly more incidence of large
positive earnings in the pre-adoption period. In terms of control variables, pre-adoption period has a higher growth than do
post-adoption period firms. Despite that there is relatively stable mean figure for size (SIZE) in both periods, the
post-adoption period is significantly larger than pre-adoption period. This implies that after moving towards IFRS, the
sample firms have grown significantly.
Table 2: Descripti ve Statistics Relating to Variables used in Anal yses for Earnings Management
Test
Vari ables
CGNI
CGCFO
ACCI
CFO
SPOS

OBS
1118
1096
1103
1109
1127

TURN
ISSUE
GROWTH
SIZE
LEV

1125
1121
1090
1097
1129

Pre-Adoption of IFRS
Post-Adopti on of IFRS
Mean
Std Dev
OBS
Mean
Std Dev
0.0091404
0.3457958
1127
-0.0005137
0.0193371
121785.4
476322.6
1122
214270.5
0.0193371
-0.0896331
2.186595
1208
8.91587
311.8548
109548
491648.4
1208
220805.8
769859.9
0.0470275
0.2117917
1208
0.0347682
0.1832681
Control Variables
1.420155
9.050101
1206
1.02097
0.9810828
0.4772906
7.697753
1128
0.1393949
0.9450538
3.053665
65.63545
1087
0.3143829
2.753595
11.12325
2.733846
1162
12.28825
2.58642
3.8324
23.26463
1208
2.952644
17.33324

Vari able Definiti on

CGNI = change in annual net income scaled by total assets
CGCFO = change in annual net cash flows fro m operating activ ities scaled by total assets
ACC1 = net income less cash flow fro m operating activit ies scaled by total assets;
CFO = annual net cash flow fro m operating activit ies divided by total assets ;
SPOS = dummy variable that equals 1 for observations for which annual net income is scaled by total assets, that
is between 0 and 0.01, and 0 otherwise;
TURN = sales divided by total assets;
ISSUE = percentage change in common stock;
GROWTH = percentage change in sales;

www.tjprc.org

editor@tjprc.org

10

Benjamin Yeboah & Michael Ye boah

SIZE = natural logarith m of sales; and

LEV = total liabilit ies divided by total shareholders equity.
It is not surprising to find that the leverage ratio (LEV) and the percentage change in total liabilities (TURN) have
decreased following IFRS adoption. The adoption principles restrict both short and long term borrowings.
Table 3 reports descriptive statistics for pre-adoption and post-adoption of IFRS firms of JSE market by industry
classification as adapted from Barth et al (2009). Table 3 suggests that a large
Table 3: IFRS Firm Sample Composition by Industry Classification
Industry Classification
1.Finance
2.Services
3.Extraction
4.Merchandise
5.Manufacturing

Pre-Adoption of IFRS
Frequency Percentage
Cum
105
8.29
8.29
696
54.93
63.22
140
11.05
74.27
224
17.68
91.95
104
8.05
100.00
1269

Post-Adopti on of IFRS
Frequency Percentage Cum
105
8.29
8.29
699
55.17
63.46
147
11.60
75.06
231
18.23
93.29
85
6.71
100.00
1266

number of firms are within the service industry classificat ion. Finance industry classification recorded the small
number of firms, wh ile as merchandise industry classification records the second largest number of firms in this industry.
We do not conduct significance tests of differences in mean between the two periods. Industry year effects were included
when constructing our metrics.
2. Earnings Management
Table 4 presents results comparing the quality of accounting amounts for IFRS and South African GAAP of firms
listed in the JSE in the post-adoption period. It reveals that post-adoption period of IFRS generally evidence less earnings
management of accounting amounts than do firms applying national standards.
Table 4: Earnings Management Results
Earnings Management Metric
Measure
Predicti on
POST
Variability of CGNI
Post>Pre
2.980
Variability of CGNI over CGCFO
Post>Pre
2.576
Correlation of ACC1 and CFO
Post>Pre
0.0544
Small positive NI (SPOS)
-

PRE
2.332
1.692
0.0456
0.4694

1. Variability of Change in Net Income

This measure of earn ings management indicates that post-adoption of IFRS depicts a significantly higher
variability of change in net income (CGNI), of 2.980 co mpared to pre-adoption period CGNI measure of 2.332
(see Table 4). This suggests that there is higher variability in earnings in the post -adoption period by 0.648. The higher
variability of net inco me in the post-adoption period implies that, IFRS tends to min imize the ability of managerial
discretion over accounting numbers, which in turn leads to improved accounting quality as consistent with Barth et al.
(2008).
2. Variability of the Rati o of CGNI and CGCFO
This measure expresses the ratio of the variance of the change in net income to the variance of the change in
Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0

11

When is Accounting Quality Improved? Assessing the Earnings Management Effect of IFRS for South Africa

operating cash flows. This ratio is used to ensure that the volatility of net inco me is not driven by volatility in the operating
cash flow (Barth et al, 2008; Paananen, 2008) Table 4 shows that post-adoption period ratio (2.576) is substantially higher
than in the pre-adoption period ratio (1.692). It suggests that the variability of the change in net income in the pre-adoption
period is below the variability of the change in operating cash flows. This implies that there is less smoothing of earnings
to cause an improvement in accounting quality than it is more likely by the effect of accruals, which arise fro m the IFRS
adoption (Barth et al, 2008).
3. Correlati on of ACC1 and CFO
This measure shows the correlation between accruals (ACC1) and cash flows (CFO) variables. It depicts
significantly more positive in the post-adoption period (0.0544) than in the pre-adoption period (0.0456) (see Table 4).
It suggests that earnings smoothing has reduced as a result of the adoption of IFRS , which in turn, leads to improve the
accounting quality (Chua et al, 2012). This suggests the notion that the IFRS adoption has lowered earnings management
by way of smoothing the significant coefficient for post-adoption period, as adoption limits managers ability to exercise
discretion on accounting numbers to their short-term advantages.
4. Small Positi ve Net Income towards Target
The coefficient of SPOS (0.4694) is positive, although it is not statistically significant. This is contrary to the
prediction that IFRS adoptions generally improve accounting quality. This imp lies that there is insignificant difference in
terms of firms managing earnings towards a positive target across the pre-adoption and post-adoption periods.
This suggests a higher occurrence of small positive earnings in the post -adoption period of under IFRS. This is consistent
with Barth et al (2008) studies, which stipulates that a positive coefficient between the pre and post adoption period
indicate that International Accounting Standards (IAS) firms manage earnings towards small positive amounts more
frequently in the post-adoption period than they do in the pre-adoption period. This finding is therefore consistent with the
prediction that earnings are managed more in the post-adoption period of under IFRS.
5. Discretionary Accrual Method
This study uses non-discretionary accruals of the cross -sectional modified Jones model, which is augmented to
control for the impact of firm performance on accruals by recognizing the current year return on asset (ROA)
(Kothari et al, 2005;Sam et al, 2010; Chen et al, 2010) and for growth purposes (McNicholas, 2002). Th is method is used
to assess the extent of earnings management in absolute value terms (Aboody et al, 2005). Consistent with our prediction,
post-adoption mean measure (-7.09e-11) is far less than the pre-adoption period mean measure (2.47e-10) (see Table 5).
This signifies that there is an improved accounting quality in the IFRS adoption period than in the pre -adoption period
(Chen et al, 2010).
Table 5: Discretionary Accrual Results
Discretionary
Accrual Method
PRE: racc1
POST: racc1

Predicti on: Post < Pre

Mean
2.47e-10
-7.09e-11

Standard Deviation
0.1873346
0.2311225

50%
0.0109178
0.100541

75%
0.617543
0.0631612

OBS
635
870

Post-adoption period has a larger med ian value than that of the pre-adoption period, which is a reverse of mean
values.
www.tjprc.org

editor@tjprc.org

12

Benjamin Yeboah & Michael Ye boah

Tables 6a and 6b provide a Spearman correlation mat rix results between IFRS adoption and discretionary accruals
for both pre-adoption period and the post-adoption period.
Table 6a: Pre-Adoption Peri od: S pearman Correlation for Discretionary Accruals
obs = 635
racc1

racc1
1.0000
-0.0753
inta
(0.0578)
0.0044
di ffrev
(0.9127)
-0.0631
PPEta
(0.1124)
-0.3437*
ROA
(0.0000)
*significance at 0.05%

inta

Diffrev

PPEta

ROA

1.0000
-0.0537
(0.1763)
-0.2401*
(0.0000)
-0.0764
(0.0543)

1.0000
0.0743
(0.0612)
0.0046
(0.9078)

1.0000
0.0216
(0.5871)

1.0000

Table 6 b: Post-Adopti on Period: S pearman Correlation for Discretionary Accruals

Obs= 870
racc1

racc1
inta
Diffrev
1.0000
-0.0948**
inta
1.0000
(0.0511)
0.0415
-0.0470
di ffrev
1.0000
(0.2219)
(0.1664)
-0.1624*
-0.2077*
0.0792**
PPEta
(0.0000)
(0.0000)
(0.0194)
-0.4115*
-0.0861*
-0.0387
ROA
(0.0000)
(0.0111)
(0.2541)
*significance at 0.05%, ** significance at 0.010%

PPEta

ROA

1.0000
-0.0093
(0.7836)

1.0000

Vari able Definiti on

RACC1 = residual of the variables used for discretionary accrual and dependent variable;
INTA = natural log of one div ided by the total assets;
DIFFREV = change in receivables scaled by total assets,
PPEta = gross of property, plant, and equipment divided by total assets;
ROA = return on assets
Overall, correlations between the variables in both periods suggest a mean variance inflat ion factor (VIF) of 1.52,
which mu lticollinearity is not a substantive issue. The coefficients between the Property, Plant, and Equip ment (PPEta) and
natural logarith m of inverse of total assets (inta =-0.2077), and return on assets (ROA) and residual of variables involved
(racc1 = -0.3437), are found to be negatively correlated in both the pre-adoption and post-adoption periods respectively
(all at 0.05% significance level). These results are consistent with the prio r expectation that the negative correlation reflects
the natural outcome of discretionary accrual accounting, which is a signal for an imp roved acc ounting quality
(Barth et al, 2008; Leuz et al, 2003). Table 6b also depicts a positive correlation at 0.10% significance level in
post-adoption period. This positive relationship is also expected, given that frequent turnaround of firms assets tends to
improve the earnings capacity level of the firm. This in turn leads to improve the accounting quality.

Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0

13

When is Accounting Quality Improved? Assessing the Earnings Management Effect of IFRS for South Africa

5. CONCLUSIONS
This paper examines as to whether the quality of nancial reporting has increased after the mandatory a doption of
IFRS in 2005 of South African listed firms. More specically, we investigate whether the application of IFRS in listed
firms of JSE is associated with less earnings management of accounting numbers.
Our results posit that there has been some improvement in accounting quality between the pre-adoption and
post-adoption periods of IFRS. In particular, we reveal that listed South African rms exh ibit an increase in the
accounting-based attributes such as increase in variability of change in net income, increase in variability of change in net
income over a change in cash flow, and improvement in correlation of accruals and cash flow, with exception of the small
positive net income towards target (SPOS) metric, which exh ib ited a decreased in net income after IFRS adoption.
Interestingly, these findings are more pronounced for the rms in countries where the distance between the pre -existing
national GAAP and IFRS was important. These results are in line with previous studies (Barth et al, 2008; Chua et al,
2012), as such outcome is also of great interest to the IASB and other African countries that are moving to adopt the IFRS.
They also complement research into other aspects of accounting quality such as management disclosures
(Leuz et al, 2003).
Overall, our findings offer some support to the EUs decision to harmonize standards by moving to mandatory
IFRS reporting. These findings are potentially useful for accounting standard setters, capital market regulators, and
decision-makers in both developed and developing nations including African countries, and those that have not yet decided
to adopt IFRS.
Furthermore, our findings are related to a post-IFRS adoption period that is relatively longer time period as
compared to a relatively short period of most previous studies about the topic. Finally, we used seven indicators as proxies
for accounting quality. There must be further research to examine other metrics of accounting quality, such as
comparability, predictability, value relevance and persistence.
Nevertheless, among some limitat ions of the study include timely revision to the standards so as to meet the
fast-changing economic environ ment that the firms are operating. This provides invaluable opportunities for futur e
research to recognize such changes, prior studies have not been able to ascertain whether accounting quality metrics
absolutely measure accounting quality per se or the change of policy. This is because a mult i-d imensional concept of
accounting quality should recognize and address mu ltiple attributes of accounting quality rather than what prior research
looked at. We mitigate the confounding effects of other factors by including a number of control variables that have been
identified by prior studies.
Our results provide evidence to support for the adoption of IFRS, although there is still room and scope for future
research to expand this study in the areas of other accounting quality metrics.

REFERENCES
1.

Aboody, D, Hughes, J, Lau, J. (2005). Earn ings Quality, Insider Trading, and Cost of Capital, Journal of
Accounting Research 43, p. 651-673

2.

Armstrong, C. S, Guay, W. R, and Weber, J. P, (2010a). The role of informat ion and nancial reporting in
corporate governance and debt contracting. Journal of Accounting and Economics, 50 (2/3), 179234.

www.tjprc.org

editor@tjprc.org

14

Benjamin Yeboah & Michael Ye boah

3.

Barth, M.E, Landsman, W.R. and Lang, M .H. (2008), International accounting standards and accounting
quality, Journal of Accounting Research, Vo l. 46 No. 3, pp. 467-98.

4.

Bartov, E, Givoly, D, and Hayn, C, (2001). The rewards to meeting or beating earnings expectations. Journal of
Accounting and Economics, 33 (2), 173204.

5.

Chang, J, and Lian Sun, H. (2009). Crossed-listed Foreign Firms' Earnings Informativeness, Earnings
Management and Disclosures of Corporate Governance Informat ion under SOX. The International Journal of
Accounting, Vol. 44: pp. 132.

6.

Chen, H, Tang, Q, Jiang, Y, and Lin, Z (2010) The Role of International Financial Reporting Standards in
Accounting Quality: Evidence fro m the European Union: Journal of International Financial Management and
Accounting 21:3

7.

Christensen, H, Lee, E, & Walker, M. (2008, March). Incentives or standards: What determines accounting
quality changes around IFRS adoption? Financial Accounting and Reporting Section (FARS) Paper, Available at:
http://ssrn.com/abstract=1013054.

8.

Chua, Y. L, Cheong, C. S, and Gou ld, G.(2012) The Impact of Mandatory IFRS Adoption on Accounting Quality:
Ev idence fro m Australia: Journal of International Accounting Research Vo l 11. No. 1, pp. 119 -146

9.

Collins, D, and Hribar, P. (2002). Errors in Estimating Accruals: Imp licat ions for Emp irical Research. Journal of
Accounting Research, Vol. 40, No. 1: p105-135.

10. Daske, H, Hail, L, Leu z, C. and Verdi, R. (2008), Mandatory IFRS report ing around the world: early evidence on
the economic consequences, Journal of Accounting Research, Vol. 46 No. 5, pp. 1085 -1142.
11. Dechow, P. (1994). Accounting Earnings and Cash Flows as Measures of Firm Performance: The Role of
Accounting Accruals. Journal of Accounting and Economics 1 8: 3-42.
12. Dechow, P.M and Skinner, D.J. (2000) Earnings management: reconciling the views of accounting academics,
practitioners, and regulators. Accounting Horizons, Vol. 14, No. 2, pp. 235 250
13. Dechow, P. M, Sloan, R.G. and Sweeney, A.P. (1995). Detecting earnings management. The accounting review,
v. 70, iss. 2, pp 193-225
14. De Angelo, L, (1986) Accounting Numbers as Market Valuation Substitutes: A Study of Management Butouts of
Public Stockholders. Accounting Review 61, p. 400-420
15. Ewert, R, and A. Wagenhofer (2005). Economic effects of tightening accounting standards to restrict earnings
management. The Accounting Review 43: 11011124.
16. Goncharov, I. (2005) Earnings management and its determinants: Closing Gaps in Empirical Accounting research.
Peter Lang Gmb H, Fran kfurt.
17. Healy. P.(1985). The Effect of Bonus Schemes on Accounting Decisions. Journal of Accounting and Economics
7, p.85-107
18. Healy, M, Wahlen, M, (1999) A Review of the Earnings Management Literature and its Implications for Standard

Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0

15

When is Accounting Quality Improved? Assessing the Earnings Management Effect of IFRS for South Africa

Setting. Accounting Horizon, 13, p. 365- 383

19. Huang, H.W, Mishra, S, and Raghunandan, K. (2007). Types of Nonaudit Fees and Financial Reporting Quality.
Auditing: A Journal of Pract ice and Theory, Vo l. 26, No. 1: pp. 13345.
20. Holland, D, and Ramsay, A. (2003). Do Australian Co mpanies Manage Earn ings to Meet Simple Earnings
Benchmarks? Accounting and Finance, Vol. 43, No. 1: pp.4 1-62.
21. Jaggi, B, Leung, S, and Gu l, F. (2009). Family Control, Board Independence and Earnings Management:
Ev idence Based on Hong Kong Firms. Journal of Accounting and Public Po licy, Vo l. 28: pp. 281300.
22. Jeanjean, T, and H. Stolowy (2008). Do accounting standards matter? An exploratory analysis of earnings
management before and after IFRS adoption. Journal of Accounting and Public Po licy 27: 480494.
23. Jiang, W, Lee, P, and Anandarajan, A. (2008). The Association Between Corporate Governance and Earnings
Quality: Further Evidence using the GOV-Score. Advances in Accounting incorporating Advances in
International Accounting, Vol. 24: pp. 191-201.
24. Jones, J. (1991) Earn ings Management During Import Relief Investigations, Journal of Accounting Research 29,
p. 193-228
25. Kothari, S. (2001). Capital Markets Research in Accounting. Journal of Accounting & Economics, Vo l. 31, No. 13: pp. 105-232.
26. Kothari, S, Leone, A, & Wasley, C. (2005). Perfo rmance matched discretionary accrual measures. Performance
matched discretionary accrual measures, 39, 163-197.
27. Lang, M, J. Smith Raedy, and W. Wilson (2006). Earnings management and cross listing: Are reconciled earnings
comparable to U.S. earnings? Journal of Accounting and Economics 42 (1 2): 255 283.
28. Leu z, C, D. Nanda, and P. Wysocki (2003). Earn ings Management and Investor Protection: An International
Co mparison. Journal of Financial Economics 69: 505-527.
29. Matsunaga S, Park C. (2001) The Effect of Missing a Quarterly Earnings Benchmark on the CEOs Annual
Bonus, The Accounting Review, Vol. 76, No 3, pp. 313-332
30. McNichols, M. and Wilson, G.P. (1988) Ev idence of earn ings management fro m the provision for bad debts.
Journal of Accounting Research, Vol. 26, Supplement 1988, pp. 1 31
31. Morais, A.I, and Curto, J. D.(2009) Mandatory Adoption of IASB Standards: Value Relevance and
Country-Specific Factors: Australian Accounting Review, No 49, Vo l 19, Issue 2
32. Mulford, C. W, Co miskey, E. (2002). The Financial Nu mbers Game: Detecting Creat ive
33. Accounting Practices. John Wiley & Sons, Inc, New York.
34. Paananen, M, & Lin, H. (2009). The Develop ment of Accounting Quality of IAS and IFRS over Time: The Case
of Germany. Journal of International Accounting Research, 8 (1), 31-55.
35. Payne, J, and Robb, S. (2000). Earn ings Management: The Effect of Ex Ante Earn ings Expectations. Journal of

www.tjprc.org

editor@tjprc.org

16

Benjamin Yeboah & Michael Ye boah

Accounting, Auditing and Finance Vo l. 72, No. 4: pp. 153 186

36. Peasnell, K.V, Pope, P.F, and Young, S, (2005). Board mon itoring and earnings management: do outside directors
inuence abnormal accruals? Journal o f Business Finance & Accounting, 32 (7/8), 13111346.
37. Prather-Kinsey, J. (2006) Developing countries converging with developed country accounting standards:
evidence fro m South Africa and Mexico. The International Journal of Accounting, 41:141 -162
38. Sam, H, Kang, T, Salter, S, and Yoo, Y. K.(2010) A cross -country study on the effects of national culture on
earnings management, Journal of International Business Studies, Vo l. 41, 123-141
39. Sarkar J, Sarkar S, Sen K. (2006). Board of Directors and Opportunistic Earnings Management: Evidence fro m
India, Working Paper, Indira Gandhi Institute of Develop ment Research, Mu mbai, India.
40. Schipper, K. (1989), Co mmentary on earnings management, Accounting Horizons Vol. 3, No.4, pp.91-102
41. Watts R. and J. Zimmerman (1978), Towards a Positive Theory of the Determination of Accounting Standards,
The Accounting Review, Vo l.53 No.1, January, pp. 112-134.
42. Zeghal, D, Sonda, M, Chtourou, S.M, and Fourati, Y.M (2012) The Effect of Mandatory Adoption of IFRS on
Earnings Quality: Evidence fro m the European Union. Journal of Accounting Research, Vol.11, No. 2. pp 1 -25

Impact Factor (JCC): 4.4251

Index Copernicus Value (ICV): 3.0